Brian E Dixon1, Jason A Siegel, Tanya V Oemig, Shaun J Grannis. 1. Indiana University, School of Informatics and Computing, Indianapolis, IN ; Regenstrief Institute, Indianapolis, IN ; Department of Veterans Affairs, Veterans Health Administration, Health Services Research and Development Service, Center of Excellence on Implementing Evidence-Based Practice, Indianapolis, IN.
Abstract
OBJECTIVE: We examined completeness, an attribute of data quality, in the context of electronic laboratory reporting (ELR) of notifiable disease information to public health agencies. METHODS: We extracted more than seven million ELR messages from multiple clinical information systems in two states. We calculated and compared the completeness of various data fields within the messages that were identified to be important to public health reporting processes. We compared unaltered, original messages from source systems with similar messages from another state as well as messages enriched by a health information exchange (HIE). Our analysis focused on calculating completeness (i.e., the number of nonmissing values) for fields deemed important for inclusion in notifiable disease case reports. RESULTS: The completeness of data fields for laboratory transactions varied across clinical information systems and jurisdictions. Fields identifying the patient and test results were usually complete (97%-100%). Fields containing patient demographics, patient contact information, and provider contact information were suboptimal (6%-89%). Transactions enhanced by the HIE were found to be more complete (increases ranged from 2% to 25%) than the original messages. CONCLUSION: ELR data from clinical information systems can be of suboptimal quality. Public health monitoring of data sources and augmentation of ELR message content using HIE services can improve data quality.
OBJECTIVE: We examined completeness, an attribute of data quality, in the context of electronic laboratory reporting (ELR) of notifiable disease information to public health agencies. METHODS: We extracted more than seven million ELR messages from multiple clinical information systems in two states. We calculated and compared the completeness of various data fields within the messages that were identified to be important to public health reporting processes. We compared unaltered, original messages from source systems with similar messages from another state as well as messages enriched by a health information exchange (HIE). Our analysis focused on calculating completeness (i.e., the number of nonmissing values) for fields deemed important for inclusion in notifiable disease case reports. RESULTS: The completeness of data fields for laboratory transactions varied across clinical information systems and jurisdictions. Fields identifying the patient and test results were usually complete (97%-100%). Fields containing patient demographics, patient contact information, and provider contact information were suboptimal (6%-89%). Transactions enhanced by the HIE were found to be more complete (increases ranged from 2% to 25%) than the original messages. CONCLUSION: ELR data from clinical information systems can be of suboptimal quality. Public health monitoring of data sources and augmentation of ELR message content using HIE services can improve data quality.
Authors: Anil A Panackal; Nkuchia M M'ikanatha; Fu-Chiang Tsui; Joan McMahon; Michael M Wagner; Bruce W Dixon; Juan Zubieta; Maureen Phelan; Sara Mirza; Juliette Morgan; Daniel Jernigan; A William Pasculle; James T Rankin; Rana A Hajjeh; Lee H Harrison Journal: Emerg Infect Dis Date: 2002-07 Impact factor: 6.883
Authors: Brian E Dixon; Zuoyi Zhang; Janet N Arno; Debra Revere; P Joseph Gibson; Shaun J Grannis Journal: Public Health Rep Date: 2020-04-06 Impact factor: 2.792
Authors: Khoa A Nguyen; David A Haggstrom; Susan Ofner; Susan M Perkins; Dustin D French; Laura J Myers; Marc Rosenman; Michael Weiner; Brian E Dixon; Alan J Zillich Journal: Appl Clin Inform Date: 2017-03-08 Impact factor: 2.342